Practice recommendations
- Use the basal/bolus insulin regimen for inpatients with diabetes. It follows normal physiological insulin rhythm and is associated with significantly better glycemic control than the sliding-scale regimen. (B)
- If a patient on a basal/bolus regimen consistently requires supplemental insulin, reevaluate baseline dosing and make adjustments as needed. (B)
- Whenever possible, switch hospitalized patients to their outpatient diabetes control regimen ≥24 hours prior to discharge. (C)
Strength of recommendation (SOR)
- Good-quality patient-oriented evidence
- Inconsistent or limited-quality patient-oriented evidence
- Consensus, usual practice, opinion, disease-oriented evidence, case series
Mr. H, a 62-year-old with type 2 diabetes, hypertension, and hypercholesterolemia, arrives at the emergency department complaining of acute onset chest pain. An EKG shows no ischemic changes and his initial cardiac enzymes are normal, but Mr. H is admitted to telemetry for further monitoring and to rule out myocardial infarction. Mr. H normally takes metformin and glipizide to manage his diabetes; his most recent glycosylated hemoglobin (HbA1c) was 8.2. After admission, he is placed on a diabetic diet and switched to insulin.
As primary care physicians, we all care for patients like Mr. H, who are hospitalized because of cardiovascular or other symptoms and have diabetes—a comorbidity that affects an estimated 12% to 25% of inpatients.1 We are also well aware of the elevated risks such patients face—for bacterial infection, impaired wound healing, and reduced tissue and organ perfusion,2 among others. In one study, a single blood glucose reading >220 mg/dL was associated with a nearly 6-fold increase in nosocomial infection.2 A number of recent studies have also found hyperglycemia to be an independent marker of overall inpatient mortality.1,3-5
American Diabetes Association goals. In 2008, the ADA issued new glycemic control goals for inpatients with diabetes. For critically ill patients, the association recommends that blood glucose levels be maintained at <140 mg/dL—and as close to 110 mg/dL as possible. For patients who are hospitalized but are not critically ill, the ADA recommends fasting blood glucose levels of 90 to 130 mg/dL and postprandial levels <180 mg/dL.6
As the ADA recommendations make clear, it is imperative that we do everything possible to lower the blood glucose levels of our hospitalized patients. Ironically, though, fear of hypoglycemia has prevented many physicians from putting patients with diabetes on a basal/bolus insulin protocol1—a dosing regimen that, according to at least one recent report, is more effective than the traditional sliding-scale insulin regimen.7 (See “No heightened hypoglycemia risk with basal/bolus regimen”) To help you achieve glycemic targets safely and confidently using the basal/bolus regimen, we’ve assembled this review of the latest evidence, complete with strategies for success.
Oral agents are no match for the hospital routine
The hospital environment interferes with the patterns and schedules that people with diabetes rely on to manage their condition. Thus, it is not unusual even for patients whose glucose levels were very well-controlled at home to have poor glycemic control as inpatients. Dietary change is one of the primary reasons.
Mealtimes typically deviate from the patient’s at-home schedule. In addition, patients are often put on a calorie-restricted, carefully enforced diabetic diet, which is quite different from their usual eating pattern. NPO orders are also common in preparation for diagnostic testing or other procedures. And some medications—particularly high doses of steroids—affect glucose levels. It is difficult to adjust oral hypoglycemic agents to accommodate such variations.
A look at Mr. H’s regimen. Mr. H’s physician knew that continuation of his oral medications—particularly glipizide—in combination with the hospital’s strict diabetic diet could result in hypoglycemia. Continuing to take metformin was also a concern, given that Mr. H was at risk for new cardiac symptoms—a contraindication to metformin use. So his physician switched him over to insulin, a safer alternative.
Finding the right insulin regimen
For years, a sliding-scale regimen was the most common approach to glycemic management of inpatients with diabetes. This concept, developed in 1934, originally used urine glucose testing to determine dosing, and its convenience and ease of treatment initiation led to widespread use. Although many variations have been introduced over the years, traditional sliding-scale regimens use short-acting analog or regular insulin in predetermined doses based on blood glucose readings at mealtimes and bedtime.
Despite the popularity of this method, however, there is little evidence to support it. Sliding-scale insulin as monotherapy has not been associated with effective glycemic control or improved outcomes.8,9 By design, this traditional regimen makes hyperglycemia the threshold for action, rather than taking action to prevent it. The result: wide fluctuations in blood sugar levels and the potential for prolonged periods of hyperglycemia.